I'd suggest that, when grouping, you use the most frequently occurring filename in the group as the value in the Filename column. When frequency for a name is tied, choose the most frequent-appearing name with the earliest returned result.
This would produce more consistent results when the majority of sharers are following a standard naming convention, but various other naming conventions are in minority use.
Using a "first result returned" method produces group names using the various naming convention based their relative prevalence. Which just looks messy.
It would be more coding to use the frequency-based approach, but it would look much slicker in most cases I see.
I'd suggest that, when grouping, you use the most frequently occurring filename in the group as the value in the Filename column. When frequency for a name is tied, choose the most frequent-appearing name with the earliest returned result.
This would produce more consistent results when the majority of sharers are following a standard naming convention, but various other naming conventions are in minority use.
Using a "first result returned" method produces group names using the various naming convention based their relative prevalence. Which just looks messy.
It would be more coding to use the frequency-based approach, but it would look much slicker in most cases I see.